Event Timeline vs Causal Map vs Barrier Review: Which Investigation Lens Fits the Evidence?

A serious incident rarely arrives as a single bad decision. The evidence is usually scattered across the sequence of work, the conditions that shaped decisions, and the controls that were supposed to prevent harm. An investigation becomes useful when its method matches the evidence rather than when the team applies a familiar diagram to every event.

An event timeline, a causal map, and a barrier review answer different questions. The timeline reconstructs what happened and when. The causal map explains how conditions and decisions connected. The barrier review tests why a control did not prevent, detect, or limit the event. Treating them as interchangeable produces a polished report that may still leave the repeat pathway intact.

Across more than 250 cultural transformation projects, Andreza Araujo has emphasized a practical distinction between describing an event and changing the conditions that made it possible. The investigation method should therefore be selected by the decision the organization must make next, not by the template stored on the shared drive.

Evaluation criteria for selecting an investigation lens

The first criterion is the evidence question. If people disagree about the order of events, the team needs a sequence that can be checked against records, interviews, access data, and physical traces. If the order is clear but the reasons remain unclear, the team needs a structure that connects workload, design, supervision, competence, and local decisions without reducing the event to one person.

The second criterion is the control question. Leaders need to know whether a safeguard was absent, poorly designed, unavailable, defeated, or simply never verified. That question cannot be answered by chronology alone, because a timeline can show that a permit existed without showing whether the permit controlled the exposure.

Use five dimensions when choosing the method. Compare the clarity of the sequence, the complexity of the contributing conditions, the visibility of failed controls, the quality of available evidence, and the decision that the investigation must support. The following matrix keeps those dimensions visible before the team starts drawing.

The methods can be combined, but they should not be blended into one undifferentiated exercise. A timeline often establishes the facts first, a causal map then shows the relationships, and a barrier review identifies which changes deserve priority. The sequence matters because a control judgment built on an unverified event order can send the organization toward the wrong corrective action.

When an event timeline is the right choice

An event timeline is the strongest starting point when the team does not yet agree on what happened. This is common after a line stoppage, an unexpected release, a vehicle collision, or a medical emergency in which several people saw different parts of the work. The method creates a shared frame without asking the team to explain every condition too early.

Build the timeline from observable events, not interpretations. Place the work permit, shift handover, alarm, equipment state, communication, intervention, and response in sequence. Add the source for each entry, because a statement from an interview should not carry the same certainty as a timestamped control-system record.

The timeline becomes more valuable when it marks decision points. Identify where someone chose to continue, pause, escalate, isolate, restart, or accept a deviation. Those points show where the work moved from a planned state to a changed state, which is often where the organization needs better control ownership.

James Reason's work on latent conditions supports this discipline because it separates the visible action from the organizational conditions that shaped it. A timeline does not prove blame. It gives the team a reliable sequence from which deeper questions can be asked.

The main limitation is that chronology can create false confidence. A complete sequence may still fail to explain why a warning was ignored, why a procedure did not fit the task, or why supervision accepted a weak signal. When the sequence is stable but the decision logic is not, move to a causal map.

When a causal map explains more than chronology

A causal map is appropriate when several conditions interacted and no single event explains the outcome. It can connect staffing pressure, production change, equipment condition, planning quality, competence, supervision, and local workarounds, provided each connection is supported by evidence rather than by the team's preferred story.

Start with the unwanted outcome, then work backward through immediate conditions and enabling conditions. Ask what had to be true for the exposure to exist, what allowed the exposure to continue, and what made the final action seem reasonable at the time. The map should show relationships, not merely collect every weakness discovered during interviews.

A useful map distinguishes evidence from inference. A maintenance log can demonstrate that a defect remained open. An interview may explain how the defect affected the task. A hypothesis that the defect caused the incident requires further support. Keeping those levels separate protects the investigation from turning uncertainty into a confident narrative.

The method is especially useful when the event reflects a work-design problem. If overtime, conflicting priorities, an unclear escalation route, and an unreliable alarm all appear in the evidence, a causal map can show how they combined without pretending that one factor was the sole cause.

The main limitation is speculative expansion. Maps become unhelpful when every plausible factor is added, because the team loses the difference between a proven contributor and a general concern. Each branch needs a source, a decision implication, and a clear boundary around what remains unknown.

When a barrier review should lead the investigation

A barrier review should lead when the central management question is whether a critical safeguard worked. This is the right lens for high-energy releases, loss of containment, serious vehicle events, falls, unexpected movement, and other exposures in which a small number of controls should have interrupted the pathway.

List the expected barriers across prevention, detection, and mitigation. Then test each one against the work as performed. Was the barrier present? Was it available at the point of exposure? Did the person understand its purpose? Could it be bypassed? Was its condition checked? Did someone own the decision to restore it after a temporary change?

The review should examine both technical and administrative controls, although the evidence standard must remain practical. A procedure is not a functioning barrier merely because it is approved. A training record is not proof of task competence. A field inspection is not evidence of control if it did not examine the critical failure mode.

Barrier review also helps leaders assign corrective-action ownership. Instead of writing “reinforce awareness,” the team can specify that the isolation verification must be redesigned, the alarm test must be recorded at the point of use, or the supervisor must have an escalation trigger when the planned control is unavailable.

The main limitation is narrowness. A barrier review can identify a weak safeguard while missing the workload, planning, or leadership conditions that allowed the weakness to persist. Pair it with a timeline or causal map when the investigation must explain both control performance and organizational conditions.

Decision matrix: match the method to the evidence

Choose the event timeline when facts are disputed, records are time-sensitive, or the team needs to establish the sequence before debating causes. Choose the causal map when the sequence is mostly known but several conditions shaped the exposure and the organization must understand their interaction.

Choose the barrier review when the event involved a critical hazard and the next decision concerns control reliability. If a serious event could recur because a safeguard is unavailable, bypassed, poorly tested, or owned by nobody, the investigation must produce more than a narrative. It must identify the control test that will change the exposure.

For a serious event, the most reliable sequence is usually chronological, causal, then control-focused. The team first establishes what happened, then explains the conditions that shaped the work, and finally tests which barriers should change. This order reduces the risk that a familiar corrective action will be selected before the evidence is understood.

Recommendation for investigation leaders

Do not ask which method is universally superior. Ask which uncertainty is blocking a safer decision. If uncertainty concerns the order, begin with the timeline. If it concerns interaction, build the causal map. If it concerns the reliability of protection, lead with the barrier review. When the event is consequential and complex, use all three in sequence rather than forcing one tool to carry the whole explanation.

Every investigation should finish with a short statement that a supervisor can use. Name the exposure, the failed or missing control, the evidence supporting that judgment, the owner who can change it, and the test that will show whether the risk actually moved. That is how an investigation becomes a management control rather than a document that closes an action list.

Andreza Araujo's approach to safety culture keeps the standard practical. Across 25+ years in multinational EHS leadership, she has seen that learning becomes operational only when the organization changes a decision, a control, or the conditions under which work is performed. The investigation lens is valuable when it produces that change.

Explore Andreza Araujo's work on safety culture, leadership, and risk decisions.